Analysis of IEC 61850-9-2LE Measured Values Using a Neural Network
- Section 2: Time synchronization over a process bus. This section contains the test structure with the devices used during the test. It contains the GPS parameters and initial test of the signal which generated from the GPS.
- Section 3: The IEC 61850 sampled measured values testing. This section contains the sampled values test with the OMICRON device and the test structure and sampled values directions.
- Section 4: The timing analysis of sampled values streams. This section contains the result of the measurement of the OMICRON merging unit and physical relays. (CMC-publisher, IED-publisher- 2x IED-subscriber) when time synchronization is applied, however, the interval time is around 240 µs with the local clock of the publisher IED or CMC merging unit, the interval time is around 230 µs with the global clock (GPS is applied).
- Section 5: Generic Object Oriented Substation Events (GOOSE). A GOOSE trip signal is sent from the publisher IED to the subscriber IED. This test found that when the GOOSE message is sent to the receiver IED (tripping signal), the signal is duplicated four times with a size of 147 bytes per packet, the average interval time between the packets was practically constant from the first to the fourth packets (278 µs) and the average interval between the fourth and the fifth packet was 102 ms.
- Section 6: Machine learning. By machine learning, we found a link between two parameters (number of samples/ms – interval time) and used to determine the publisher. The inputs and the target provided to the network and the algorithm breaks up the data for test sets (training 70%—validation 15%—testing 15%), the best validation was in epoch 23.
1.1. Impact of IEC 61850 on Substation Operations
- Increase the power quality, reducing the copper wires that are plugged into the IEDs and acheiving faster response to short circuit faults.
- The IEC 61850 standard provides the interoperability between IEDs from various manufacturers and offers a friendly configuration that can be implemented at a site without external support.
- Reducing the cost of operation and maintenance in the substations.
- This standard provides secure and fast data transmission among IEDs and substation devices.
- The IEC 61850 functionality is flexible and easy to implement by using the tools properly.
- The IEC 61850 configuration should follow the performance requirements.
- The communication interface between IEDs and the system should follow the IEC 61850 standards communication fundamentals.
- Establish the communications between devices by mean transfer data between IEDs and the power system among the SAS.
- Data characterization such as sampling frequency, sampling counts, and time synchronization.
- Communication specification that can be summarized as a generic object oriented substation event (GOOSE), manufacturing message specification (MMS), SMV, and Ethernet communication.
- The data structures and data objects’ services.
- Definition and determine how to access the structure for the data’s abstract communication services interface (ACSI) and the configuration of the communication solution and compatible protocols.
- Standardizing the output data from IEDs and categories the sharing of data between GOOSE and SMV orders
- The IEDs and network communications are implemented using eXtendable Markup Language (XML).
1.2. The IEC 61850 Information System
2. Time Synchronization over a Process Bus
PTP Clock Types in Time Synchronization
3. The IEC 61850 Sampled Measured Values Testing
3.1. Sampled Values Test
3.2. SVScout Software
4. The Timing Analysis of Sampled Values Streams
Time Display and Time References
- While packets are captured, each packet is timestamped. Each capture file includes these timestamps which is important for later analysis.
- According to RFC 2544 testing methodology, RFC 2544 requires the standard frame sizes (128, 256, 512, and 1280) bytes.
5. Generic Object Oriented Substation Events (GOOSE)
6. Machine Learning
- IED-subscriber took time to determine the publisher merging unit, and a delay time to recognize the publisher side. Practical implementation showed that the simulated merging unit of the IED could not subscribe immediately.
- The number of samples is the first input parameter for data preparation; each merging unit includes a number of samples per second.
- Interval time between packets used in this test is the second input parameter for data preparation.
- Measurement of the merging units showed that the quick response of the merging unit subscriber is important in IEC 61850. The data link layer (layer 2) is a lower level addressing structure to be used between end systems and concerned with forwarding packets based on layer addressing scheme and the MAC address of the destination.
- The interval time and the number of samples are parameters used as inputs for this test, using relay protection merging unit and CMC merging unit data to feed the training set and test set.
- By using neural net pattern recognition, we could find the relation between the inputs (number of samples/ms—interval time between the packets) and the source of the data.
- By using this technique, technicians can save time and ensure they are testing the correct merging unit.
- According to our test the subscriber protection relay takes time to respond to the new traffic of sampled values.
- The training set is 70%.
- The validation set is 15% to prove that the network stops training before overfitting.
- The testing set is 15% and is used as an iindependent test of network generalization .
- The IED-subscriber starts to send sampled measured values once the global clock is applied (GPS - time synchronization), conversely, IED-subscriber does not send or publish SMV once the internal clock is applied from the publisher IED.
- The interval time between the samples is 250 µs according to the IEC 61850 standard, and the network analysis tool shows that four MAC addresses are available, in this case (CMC-publisher, IED-publisher- 2x IED-Subscriber) with time synchronization is applied, however, the interval time is around 240 µs with the local clock of the publisher IED or CMC merging unit, and the interval time is around 230 µs with the global clock (GPS) applied.
- The number of samples per millisecond of IED-publisher: the number of packets is not constant, the range was between 3 to 5 packets/ms, while with CMC-publisher: the number of packets is almost constant at 5 packets/ms.
- GOOSE message configuration is implemented to the IEDs (sender-receiver), the GOOSE message is sent to the receiver IED (tripping signal), the signal is duplicated four times with a size of 147 bytes per packet, the average interval time between the packets was practically constant from the first to the fourth packets (278 µs) and the average interval between the fourth and the fifth packet was 102 milliseconds.
- IED-subscriber is subscribing the SMV from the IED-publisher and CMC-publisher equally, IED-subscriber is unable to recognize who is the publisher of the SMV (IED or CMC) due to the fact the CMC-publisher has the same dataset as the IED-publisher (that is, in fact, what happened when CMC-publisher was simulating the IED-publisher). Wherefore, a model is applied to predict if the IED-subscriber would recognize which merging unit is sending the sampled values based on different attributes, to implement the approach, train a classifier using different models and measure the accuracy and compare models, using the classifier for prediction. The preparation data includes two parameters (number of samples/ms - interval time between the packets) for each publisher of SMV (IED or CMC). By using neural net pattern recognition that solves the pattern recognition problem using two layer feed networks (nprtool), the inputs and the target provided to the network and the algorithm break up the data into test sets (training 70%- validation 15%- testing 15%), and the best validation was in the 23rd epoch.
- This method can be used for optimization of testing procedures in substations where IEC 61850-9-2LE are implemented. This method can be used for shorter test preparation, to lower the cost and the method can lower the cost and help support research projects since it allows one to implement better platform and services as well as to integrate different communication protocols when necessary.
Conflicts of Interest
- Sidhu, T.S.; Pradeep, K. Gangadharan. Control and automation of power system substation using IEC61850 communication. In Proceedings of the 2005 IEEE Conference on Control Applications (IEEE CCA 2005), Toronto, ON, Canada, 28–31 August 2005. [Google Scholar]
- Kriger, C.; Shaheen, B.; John-Charly, R.-M. A detailed analysis of the GOOSE message structure in an IEC 61850 standard-based substation automation system. Int. J. Comput. Commun. Control 2013, 8, 708–721. [Google Scholar] [CrossRef]
- Štefanka, M. Application of sensors and digitalization based on IEC 61850 in medium voltage networks and switchgears. Ph.D. Thesis, Brno University of Technology, Faculty of Electrical Engineering and Communication, Brno, Czech Republic, 2016. [Google Scholar]
- Andersson, L.; Brand, K.P.; Brunner, C.; Wimmer, W. Reliability investigations for SA communication architectures based on IEC 61850. In Proceedings of the IEEE Power Tech, St. Petersburg, Russia, 27–30 June 2005. [Google Scholar]
- Sheng, C.H.E.N. Application of system networking for substation automation. Power Syst. Technol. Beijing 2003, 27, 72–75. [Google Scholar]
- Popova, S.; Iliev, S.; Trifonov, M. Neural Network Prediction of the Electricity Consumption of Trolleybus and Tram Transport in Sofia City. In Proceedings of the Latest Trends in Energy, Environment and Development, Proc. of the Int. Conf. on Urban Planning and Transportation (UPT’14), Salerno, Italy, 3–5 June 2014; pp. 116–120. [Google Scholar]
- Dolezilek, D. IEC 61850: What you need to know about functionality and practical implementation. In Proceedings of the 2006 IEEE Power Systems Conference: Advanced Metering, Protection, Control, Communication, and Distributed Resources (PS’06), Clemson, SC, USA, 14–17 March 2006; pp. 1–17. [Google Scholar]
- Ingram, D.M.E.; Schaub, P.; Richard, R.T.; Duncan, A.C. Performance analysis of IEC 61850 sampled value process bus networks. IEEE Trans. Ind. Inform. 2003, 9, 1445–1454. [Google Scholar] [CrossRef]
- Jiang, W.; Henning, S. Modeling of packet loss and delay and their effect on real-time multimedia service quality. In Proceedings of the Nossdav’2000, Chapel Hill, NC, USA, 26–28 June 2000. [Google Scholar]
- Xiang, G.; Pei-Chao, Z. Main features and key technologies of digital substation. Power Syst. Technol. 2006, 23. Available online: http://en.cnki.com.cn/Article_en/CJFDTOTAL-DWJS200623015.htm (accessed on 27 April 2019).
- Mekkanen, M. On reliability and performance analyses of iec 61850 for digital sas. Ph.D. Thesis, University of Vaasa Faculty of Technology Department of Computer Science, Vaasa, Finland, 2015. [Google Scholar]
- Goran, J.; Havelka, J.; Capuder, T.; Sučić, S. Laboratory Test Bed for Analyzing Fault-Detection Reaction Times of Protection Relays in Different Substation Topologies. Energies 2018, 11, 2482. [Google Scholar]
- Silos, A.; Señís, A.; De Pozuelo, R.M.; Zaballos, A. Using IEC 61850 GOOSE Service for Adaptive ANSI 67/67N Protection in Ring Main Systems with Distributed Energy Resources. Energies 2017, 10, 1685. [Google Scholar] [CrossRef]
- Cai, Y.; Chen, Y.; Li, Y.; Cao, Y.; Zeng, X. Reliability Analysis of Cyber–Physical Systems: Case of the Substation Based on the IEC 61850 Standard in China. Energies 2018, 11, 2589. [Google Scholar] [CrossRef]
- O’Reilly, R.; Beng, T.C.; Dogger, G. Hidden challenges in the implementation of 61850 in larger substation automation projects. 2016. Available online: https://pdfs.semanticscholar.org/02c7/2e128c2937b31474c40cb5a5bde50f06ecc5.pdf (accessed on 27 April 2019).
- Wannous, K.; Toman, P. Sharing sampled values between two protection relays according to standard IEC 61850-9-2LE. In Proceedings of the 19th IEEE International Scientific Conference on Electric Power Engineering (EPE), Brno, Czech Republic, 16–18 May 2018. [Google Scholar]
- Mlynek, P.; Misurec, J.; Toman, P.; Silhavy, P.; Fujdiak, R.; Slacik, J.; Samouylov, K. Performance Testing and Methodology for Evaluation of Power Line Communication. Elektron. Elektrotechnika 2018, 24, 88–96. [Google Scholar] [CrossRef]
- Patel, N. IEC 61850: Horizontal Goose Communication and Overview: IEC 61850 Horizontal Communication, Goose Messaging and Documentation: IEC 61850 Standard Overview and Understanding; Lambert Academic Publishing: Saarbrücken, Germany, 2011. [Google Scholar]
- Stefanka, M.; Prokop, V.; Salge, G. Application of IEC 61850-9-2 in MV Switchgear with Sensors Use; IET: London, UK, 2013. [Google Scholar]
- Raussi, P. Real-Time Laboratory Interconnection for Smart Grid Testing; LUT University: Petra, Jordan, 2017. [Google Scholar]
|Timing Accuracy||Supported Timing Protocols||Specifications||Channels and Frequency|
|±100 ns to reference time (TAI/UTC)||PTP||CMGPS 588 synchronization unit with integrated antenna and timing receiver||12 channel GPS receiver, Frequency: 1575.42 MHz|
|CMCGPS 588||Time Interval|
|GPS||Locked||Sync interval||1 s|
|PTP||Master||Announce interval||1 s|
|NTP||Synchronized||Announce receipt timeout||3 s|
|Satellites usable||4||Peer mean path delay||85 ns|
|Linear Voltage Sensor||Rogowski Current Sensor|
|Nominal voltage of OMICRON||2 V||Nominal voltage of OMICRON||150 mV|
|Output System channels||Line to Line out 1–3||Output System channels||Line to Line out 4–6|
|Min. Frequency||0 Hz||Min. Frequency||0 Hz|
|Max. Frequency||1 kHz||Max. Frequency||395 Hz|
|Frame Size [bytes]||Round Trip Latency [µs]|
|IEC 61850 -9-2||50 Hz System||60 Hz System||SMV_Publisher1||SMV_Publisher2|
|HSR redundant network||4||4||12.3Mb/s||37.7 Mb/s|
|PRP redundant network||9||8||12.3 Mb/s||87.7 Mb/s|
|T||250 µs||208 µs||5.12 Mb/s, 50 Hz||6.16 Mb/s, 60 Hz|
|Interval time||(4000 × 1)||(4000 × 1)||IED|
|Number of samples per ms||(4000 × 1)||(4000 × 1)||CMC|
|Training Set||Test Set|
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Wannous, K.; Toman, P.; Jurák, V.; Wasserbauer, V. Analysis of IEC 61850-9-2LE Measured Values Using a Neural Network. Energies 2019, 12, 1618. https://doi.org/10.3390/en12091618
Wannous K, Toman P, Jurák V, Wasserbauer V. Analysis of IEC 61850-9-2LE Measured Values Using a Neural Network. Energies. 2019; 12(9):1618. https://doi.org/10.3390/en12091618Chicago/Turabian Style
Wannous, Kinan, Petr Toman, Viktor Jurák, and Vojtěch Wasserbauer. 2019. "Analysis of IEC 61850-9-2LE Measured Values Using a Neural Network" Energies 12, no. 9: 1618. https://doi.org/10.3390/en12091618